DH as a discipline and the sub-disciplines of DH

The question of whether Digital Humanities (DH) is a discipline has had a whole moment, with many publications and discussions around questions like: What even makes a discipline? Has DH reached that status? How does discipline building work? How can we define DH in the first place?

There has also always been a certain reluctance within DH to truly define itself, maybe culminating in the fun website https://whatisdigitalhumanities.com/ that simply keeps giving you yet another definition of what DH is upon each reload. So the whole discussion has been about, on the one hand, defining what DH is and, on the other, wondering whether DH is a discipline, can ever be a discipline, has already become one, or what it would take for that to happen.

I think we have reached some general consensus on certain basics, but beyond that, things remain much less clear. And maybe that’s partly deliberate. Some have rightly observed that DH also seems to resist being fully defined. At the same time, DH always exists in relation to the humanities disciplines it draws from. So maybe it’s completely normal that it remains slippery and difficult to grasp depending on which perspective you look at it from. The sheer range of sources, materials, methods, and approaches we use is enormous. And by now, DH has so many sub-fields of its own that this brings me to a slightly different question: not whether DH itself is a discipline, but what is happening with the sub-disciplines within DH?

Defining “disciplineness”: What makes something a discipline?

Whether or not you believe DH as a whole should be a discipline, there are some relatively concrete markers of “disciplineness” that people have used in these discussions. Are there study programmes under that name? Are there professorships? Journals? Conferences? A research community? For Digital Humanities as a broader field, all of these things clearly exist. In fact, DH is now established enough that we can ask the same questions about some of its sub-fields.

Computational literary studies (CLS) is a really good example for me because there are very clear markers of disciplineness. There are professorships specifically for this area, conferences, journals, and, most importantly, a recognisable research community. That community by and large probably participates to some degree in both literary studies conferences and broader Digital Humanities conferences. But it also has spaces of its own, with a lot of specialised expertise. And that matters because expertise accumulates. Certain methods and topics have already been tried and tested. There are best practices. People don’t necessarily need to start every conversation from the beginning anymore, so the field can move on to increasingly specific questions and sub-interests.

One example I frequently encounter as something of an outside observer of CLS is work on canon formation, which sometimes comes close to my own interests because it connects to questions of data gaps and equal representation. I seem to end up reviewing papers about this quite a lot, and my impression is that it has developed into a really strong area of research. There are identifiable research trends within an already specialised sub-field. So, at least in cases like computational literary studies, I find it quite convincing to speak of an established sub-discipline.

But not every DH sub-field develops in the same way

The situation looks rather different in my own area (although, of course, we could already have a discussion about what exactly my sub-discipline is).

There is computational humanities, which I’ve talked about in various places. I (famously it seems) wrote about it in 2020, when I was a PhD student observing what was happening as CHR formed itself and thinking “wow, something big is happening here.” Looking back, I was witnessing a sub-discipline in its formation.

I’ve since given a few presentations in which I’ve tried to bring those thoughts from 2020 together with my background in history of science and science and technology studies, and to explain a bit more academically what I think was (and still is) happening there. The last one of those was at the recent DH2026.

Of course, the establishment of that one conference wasn’t the only thing that made computational humanities a thing. But it was certainly one point where a lot of these developments came together.

Maybe I’ll write another blog post about that at some point – let me know if you’d be interested but there probably is no stopping me should I decide to do so anyway 😉

More recently, however, I was invited to reflect on whether computational history of science, knowledge and technology is a sub-field. In my reflections, I ended up arguing that what we need is not just a computational history of these things, but digital and computational histories of science, knowledge and technology. It sounds like just more words but look closely, adding the “digital and” makes a big difference, I would argue.

A label doesn’t necessarily make a field

The history of science, knowledge and technology has had an interesting trajectory in discipline-building because people have prominently claimed names that could serve as labels for a movement or a field (“Computational History of Science”, e.g. in the 2019 Isis focus section). At the same time, there isn’t actually that strong of a community behind those labels. It would be completely wrong to say that nothingexists, of course. There is a lot of work and expertise out there. But unlike computational literary studies (CLS), I don’t think this has developed into a sub-field or sub-discipline in the same way.

For me, one very practical test of this “disciplineness” is: if I wanted to connect with people working in this field, where would I go? There isn’t really a journal or a conference as the obvious place I could simply go to. I would need to contact specific individuals whom I happen to know are working on digital methods within history of science, knowledge or technology. Whereas if I wanted to connect with people in digital or computational archaeology, for example, there are established places I could go. There are communities and mailing lists I could sign up for. There are structures that would allow me to find other people doing this kind of work without already having to know who they are. That, to me, is a barometer of “disciplineness” and this can be quite different depending on your subfield.

Why these sub-disciplines actually matter

I think having these sub-fields is really important, and my experiences working on the history of alchemy have made that particularly clear to me. Sometimes historians respond to computational work by saying something along the lines of: “This doesn’t really fit our journal. Why don’t you take it to a Digital Humanities journal or conference?” And, yes, I’m also presenting this work at DH conferences. But the DH researchers there don’t necessarily have the deep subject expertise in the history of alchemy that is needed to engage with every aspect of the research. At the same time, when I take the work into history of alchemy spaces, the computational methods can be so far outside what people normally do that the work becomes difficult to evaluate there as well.

I’ve experienced both ends of this spectrum in peer review. For the special issue I have recently co-edited, for example, one reviewer criticised one of the somewhat less computational contributions for essentially being a failure in a special issue on computational methods because the methods were “too simple.” There is a whole issue hiding in that response about the different valuation of “digital” versus “computational” work. What does “too simple” even mean here? Where exactly is the boundary between digital and computational research? And what makes either kind of study worthwhile or “enough”, whatever that means?

I’ve tried to write about this issue between digital and computational humanities more academically elsewhere, so that’s probably a topic for another blog post. – But then I’ve also experienced the opposite problem.

The problem of needing two kinds of expertise and constantly being expected to explain your entire field

Other reviewers, and initially some of the journal editors, expected us to explain what felt to me like the absolute basics of Digital Humanities, so that theoretically every member of the journal’s typical readership would be able to understand all of the digital methods being used. I pushed back against that. If I write a tutorial explaining all the basics, that can go on this blog. But that’s not really the purpose of a research article, and at some point it distracts from the actual argument I’m trying to make.

This isn’t meant to shame the journal or the society involved – I collaborate with them closely, I think they did a great job, and in the end we worked out an approach that worked for everyone. But the experience brought this problem into focus for me. The same journal publishes articles involving chemistry. I’m not a chemist, so I cannot fully understand all of the chemistry background in those papers. And that’s fine. So why should the digital methods have to be explained to the point that someone who hasn’t followed developments in Digital Humanities over the last twenty years can understand every aspect of them? There is an asymmetry there and I wonder why that is. Ultimately, experiences like this have convinced me that we do need these sub-disciplines. When they don’t exist, as I would argue is currently the case in the history of science, knowledge and technology, it makes doing this kind of work unnecessarily difficult.

There aren’t really publication venues where you can assume that people have both kinds of expertise.You’re always missing one side. In one venue, people understand your subject matter but not your methods. In another, people understand your methods but don’t have the deep domain knowledge needed to fully engage with what you’re doing. The result can be that you almost have to publish everything twice, under different perspectives, for the different communities you’re trying to serve. And still, nobody ever quite understands the whole thing.

I’ve seen just how much of a difference it makes when people do have both kinds of expertise. I’ve had the opportunity to participate in small workshops in Oxford this year and, the year before, in Abu Dhabi with people who combine expertise in my relevant historical areas with digital and computational methods. The conversations we were able to have there were simply on a different level. Those are conversations I can’t really have in a DH-only space, because too much of the specialised historical knowledge is missing. But I also can’t have them in a history-of-science-only or history-of-alchemy-only space, because too much of the methodological background is missing. When you bring together people who share both, you don’t have to constantly translate one half of the research for the other half of the room. You can actually move the conversation forward. That’s why these sub-disciplines matter.

Sub-disciplines don’t appear just because we name them

So perhaps one useful distinction is between naming a sub-field and actually building one. In the history of science, labels around computational approaches were claimed relatively early. But that doesn’t automatically mean that a computational history of science became a field.

A few people doing important work under a particular label aren’t necessarily the same thing as a research community. For me, that community-building aspect is essential. There needs to be somewhere for people to find each other. There need to be conversations that can accumulate rather than constantly starting from scratch. There needs to be enough shared expertise that work can be evaluated on its own terms rather than always having to explain one half of itself to the other.

That is where I currently see digital and computational histories of science, knowledge and technology: there is plenty of work, and there are people doing it, but the sub-discipline itself isn’t fully formed yet. And maybe looking at these smaller formations within DH also gives us another way of thinking about the old question of whether Digital Humanities itself is a discipline. Rather than only looking for definitions, we can look at what actually happens when disciplinary structures form: communities develop, expertise accumulates, shared conversations become possible, and eventually there are places where people know where to go to have them.

If you want to read more about my thoughts on digital and computational histories of science, knowledge and technology specifically, I invite you to read that epilogue I wrote for our special issue.

I’d be interested to know whether you’ve observed something similar in your own sub-discipline. Are there spaces where the subject expertise and the digital or computational expertise actually come together or are you also constantly moving between communities that only understand one half of what you do?

Only now that I feel these “trading zones” forming in my sub-field do I notice how much I have missed out before by not having them.

I have a few more related thoughts coming up, so see you next time and until then,
thanks for all the fish!